[Rebuild] Communication to data consumers wrt the licence change (draft)

[errt at gmx.de]

I think your mixing up two things. One is the decision between a 
hard-cut (as 'take down DB, run changeover, put up DB') and soft-cut (as 
'make changes on the running DB and have as much time as we need'), the 
other one is meeting the April 1st deadline.

For the first, while the decision was made for a soft-cut and I don't 
want to throw over that and don't think it really has been dropped by 
now, I still think a hard-cut is LESS error-prone for one simple reason: 
It can easily be rolled back. We can check the (not-yet online) DB after 
running the process and see whether it seems to be undamaged and roll 
everything back if not. Even if after a few days a mapper pointed out 
some major error, we can still see whether it's possible to fix that 
easily or if it would be better to lose that few days of mapping, take 
the before-changeover dump and rerun the fixed process. In a soft-cut, 
rolling things back might be more difficult, as changes by mappers 
happen in between, but it has the advantage of being able to see errors 
earlier, hopefully.

For the second, regardless of whether we do a hard or a soft cut, it 
will always be a bad idea to rush things just to be done by April 1st. 
You might argue we have a few additional days if we do the soft-cut as 
decided earlier, but still there's only one week left until April 1st 
and we should consider taking a few days or weeks of extra time so 
everything is well prepared for the time we actually start the 
changeover. Not taking into account that a few days more time also means 
some hundred more agreeing mappers and some thousands of objects that 
not have to be cleaned at all.

errt

Am 23.3.2012 14:50, schrieb Frederik Ramm:
> Hi,
>
> On 03/20/2012 10:37 PM, Simon Poole wrote:
>> I've reworked the text to take in to account that we are now planning 
>> for a
>> hard cut over.
>
> I would still like to find out who exactly the "we" in this is. 
> Because I surely am not, and I am not aware of any discussion within 
> the rebuild group where we said "oh, let's change our minds".
>
> We all thought that a soft cut over would make sense when we had our 
> telephone call. Why the sudden change? Everyone I spoke to thinks that 
> the soft cut over is prudent, easier to monitor, better for avoiding 
> mistakes, and it is clear that it is less stressful.
>
> It was Simon who suddenly claimed that he "thought" the idea had been 
> "dropped". But until now I have found nobody who actually dropped the 
> idea. Simon, maybe you can find out exactly what led you to believe 
> that the idea was "dropped".
>
> I am relatively sure that the "hard cut over" is going to be a major 
> cock-up and if I cannot prevent it, then at the very least I want to 
> be able to point the finger at someone afterwards and say: This 
> person/these people have decided that we will not go the soft route.
>
> As things currently are, everyone talks as if it had somehow been 
> decided that the hard route was necessary but NOBODY seems to have 
> made this decision.
>
> I have the feeling that some of you may be thinking that it has to be 
> done this way because some people on the OSMF board are throwing a fit 
> at the thought of not getting things done by 1st April. And then, when 
> the shit hits the fan, the very same board members will say: "We 
> trusted that rebuild group/engineering group/whatever group would do 
> this in the right way, had we known that this would lead to two weeks 
> without any mapping then of course we would have acted differently...".
>
> The "hard cut" way is the more error-prone, more problematic way. 
> Being forced to run a few tests quickly over the weekend on software 
> that isn't even finished today, make a guess of how much 
> read-only-time or downtime would be required and then act on that is a 
> very high-risk game and it is NOT GOOD and NOT RIGHT.
>
> This doesn't need to drag on for months but nobody, really nobody, is 
> served by rushing it. "1st April" is a date picked out of thin air by 
> three or four people. It is not our fault if that date doesn't match 
> reality, and it is our duty to do the best for the project, not the 
> best for the ego of three or four board members.
>
> And I still want to know who exactly decided to "drop" the soft cut 
> over. I want this recorded. Who, when, made the decision that we need 
> to take the more error-prone, more stressful path that tries to get 
> the change through more quickly at the expense of everyone who wants 
> to map.
>
> Bye
> Frederik
>